Quarkonia Studies in PbPb Collisions by the ATLAS Experiment at LHC

1. Quarkonia Studies in PbPb Collisions by the ATLAS Experiment at LHC Alexandre Lebedev, Iowa State University for the ATLAS Collaboration A.Lebedev Quark Matter 2008 02/10/2008 1

2. Why Quarkonia? Quarkonium dissociation due to color screening is considered as a promising signature of QGP formation. - Different quarkonia states are expected to “melt” at different temperatures. Recent RHIC results point to importance of recombination of quarkonia in the later stages of the collisions. - Also need to consider feed-down from higher resonances Complicated picture: - It is important to measure simultaneously different quarkonia states in order to understand heavy ion collisions In this talk we study possibility to measure charmonium (J/y) and bottonium (U) states via di-muon channel in PbPb collisions by the ATLAS experiment at LHC. A.Lebedev Quark Matter 2008 02/10/2008 2

3. Measuring quarkonia in ATLAS Quarkonia are measured in di-muon decay channel. ATLAS has excellent muon detection capabilities for |h| < 2.6 and PT > ~2.5 GeV/c MDT: Monitored drift tubes (barrel and endcaps) CSC: Cathode strip chambers (endcaps) RPC: Resistive Plates Chambers (barrel trigger) TGP: Thin Gap Chambers (endcaps and barrel trigger) A.Lebedev Quark Matter 2008 02/10/2008 3

4. Algorithm U with flat PT and h distributions, weighted with Pythia distributions. generate single U simulated PbPb Hijing events simulate single U merge single U to simulated PbPb Hijing events digitize single U reconstruct single U reconstruct merged events analyze results and fill histograms Analyse ESD/AOD and produce AAN analyze results and fill histograms compare A.Lebedev Quark Matter 2008 02/10/2008 4

5. U acceptance (singles) Projections are done with PT and h distributions from pythia. U was considered to be in acceptance if both muons produced hits in muon detector. A.Lebedev Quark Matter 2008 02/10/2008 5

6. Single U (baseline for PbPb) Integrated (over PT and h) mass resolution is 177 MeV. Integrated acceptance times efficiency is ~0.19 A.Lebedev Quark Matter 2008 02/10/2008 6

7. Improving mass resolution Restrict muon pseudorapidity - but loose statistics. Best mass resolution ~111 MeV (~10% improvement) Tighter reconstruction cuts can slightly improve this number - still further loss in statistics A.Lebedev Quark Matter 2008 02/10/2008 7

8. Mass resolution (PbPb) Integrated (over PT and h) mass resolution is 192 MeV Weighted with PT distribution from pythia Intergated over PT and h centrality dependence PbPb singles A.Lebedev Quark Matter 2008 02/10/2008 8

9. Reconstruction efficiency vs centrality - Integrated (over PT and h) acceptance times efficiency is 0.12 - Factor of ~2 loss in most central collisions A.Lebedev Quark Matter 2008 02/10/2008 9

10. Muon triggers ATLAS Trigger system: - Level 1 (L1) – configurable hardware Higher Level Triggers (HLT): - Level 2 (L2) – software, relies on input from L1 - Event Filter (EF) – off-line algorithms and data model In PbPb interaction rate is expected to be ~3kHz Probably no need for L1 trigger for data taking But need HLT for analysis We studied L2 trigger efficiency using merged Upsilon/PbPb events, and fake rate using pure PbPb Hijing events. A.Lebedev Quark Matter 2008 02/10/2008 10

11. Triggering on Upsilons efficiency = (U reconstructed & muon trigger fired)/(U reconstructed) No fakes in ~250 PbPb Min. Bias Hijing events (without U merging) Any L2 muon trigger fired Trigger threshold 6GeV A.Lebedev Quark Matter 2008 02/10/2008 11

12. Background and yields (1/3) Total pp cross-section at 5.5 TeV: ~100mb bbar cross-section at 5.5 TeV: ~100mb U cross-section: ~100nb - Upsilon cross-section in pp collisions was studied using pythia: http://dprice.web.cern.ch/dprice/work/oniumvalidation-jun06.pdf - 34nb with default trigger cuts, ~150nb with relaxed trigger cuts. One Upsilon reconstructed in ~1M pp events PbPb expected luminosity is 4 1026 cm-2s-1 (Letter of Intent) - Interaction rate ~3000 Hz from Glauber calculation (David d'Enterria, nucl-ex/0302016): - Total PbPb cross-section: 7.7b - Number of binary collisions: 400 (MB); 1670 (central) Assume that both high pT muons and U scale with number of binary collisions. Then background will scale as a square of signal, and S/sqrt(B) ratio will be the same for pp and PbPb. A.Lebedev Quark Matter 2008 02/10/2008 12

13. Background and yields (2/3) Sources of background: a) charm b) beauty c) hadron decays (mostly pi & K) d) hadron punch-through Calculating background: - use pythia to get muons from charm/beauty (a and b) - generate single pions and kaons, run full simulation, plot reconstructed muon spectrum (c and d) This gives us muon spectra and multiplicities Add Upsilons to this mix. Scale everything with number of binary collisions Run MC simulation and produce di-muon invariant mass spectra. A.Lebedev Quark Matter 2008 02/10/2008 13

14. Background and yields (3/3) This plot corresponds to 24 days of running, at 3 kHz collision rate Acceptance and efficiency corrected Barrel only (|h|<1), 120 MeV U mass resolution ~15k reconstructed U1S A.Lebedev Quark Matter 2008 02/10/2008 14

15. Charmonium (J/y) Main problem: low acceptance due to minimum muon PT ~2.5-3.0 GeV Two methods considered: - both muons fully reconstructed - “tagging method” for one muon (allows muon reconstruction down to 1.5GeV). “tagging” method Mass resolution 68 MeV ~100k J/y per month, tagging method full reconstruction A.Lebedev Quark Matter 2008 02/10/2008 15

16. Conclusions and outlook - Reconstruction efficiency is reasonably good even in most central PbPb events - Mass resolution is almost unaffected in PbPb collisions - Mass resolution is good enough to separate different U states at least in the barrel region - We should be able to see U and J/y peaks in a few weeks of running - Quarkonium study in e+e- channel is underway - cC study is underway A.Lebedev Quark Matter 2008 02/10/2008 16